We can never be sure about what happened in the past. Here's why:
1. Incomplete information. For the most part, all we have is what random people decided to write down and other random people decided to preserve. Sometimes historians consider other types of evidence: memories of the still-loving, monuments, art work, left-behind scraps dug up by archeologists. Altogether it's still just a tiny remnant of all that came before in the human experience.
2. Too much information (for one person to process). Thus, a key skill of the historian is to make judgements about significance and to decide what evidence to ignore and what evidence to examine closely. History, said the Swiss historian Jacob Burckhardt "is the record of what one age finds worthy of note in another." Not everything is worthy of note in the eyes of the discerning historian. See chapter 3 of Thinking Historically: Educating Students for the 21st Century by Stephane Levesque: "What is Important in the Past: Historical Significance."
3. Subjectivity: biases, examined assumptions, blind spots. The human mind (including yours!) is prone to all kinds of distortions.
4. Conflicting evidence. Every time you think you know what happened, you find a piece of evidence that contradicts it.
5. Human beings are inscrutable: they don't always mean what the say or say what they mean. So you can't even assume that what they wrote down is true. We can know what a person said or wrote, but not what they thought or believed. Never assume the public statements of political leaders reflect their true beliefs. Always consider what they want their audience to think and do. Consider every utterance as an action, what J.L. Austin called a speech act: "almost any speech act is really the performance of several acts at once, distinguished by different aspects of the speaker's intention: there is the act of saying something, what one does in saying it, such as requesting or promising, and how one is trying to affect one's audience."
Thus, you should approach the study of history with humility and tentative conclusions and:
6. Use Hedging language: "it seems possible/probably/likely that . . . "
7. Be "voraciously open to contrary argument" (and evidence). If you think you've got it figured out, do more research.
For more on these ideas, see my Terms for Thinking Historically. Many of those terms deal with the problems identified here.